RER competitiveness and the cost of protectionism: The case of Indonesia Massimiliano Calì (World Bank) and Milan Nedeljkovic (World Bank) One of the most prominent effects of commodity booms is the appreciation of the Real Exchange Rate (RER), which decreases the relative competitiveness of the non-booming tradable sector. We document that the experience of Indonesia during the first decade of 2000s is no exception with the RER (vis-à-vis the US) appreciating over 80% between 2002 and 2011. We show that this pattern has also been in line with other commodity exporting countries as Brazil and South Africa. This RER appreciation played a major role in the relative decline of Indonesian non commodity exports, manufacturing in particular, during that period. The RER appreciation has only been partly reversed during the dramatic decline in commodity prices over the past 4 years, and it has been more sticky downwards than other commodity exporting countries. This has contributed to the sluggish performance of the non commodity exports in the post-commodity boom period. We investigate the reasons behind this stickiness in the Indonesian RER by performing a novel decomposition of the bilateral RER across different dimensions. The results show that Indonesia’s RER stickiness in 2014-15 is mainly due to local retail prices rising faster than border prices (relative to the US and Asian peers). We hypothesize that this is due to increases in tariffs and non tariff barriers as well as in distribution costs, which in turn is consistent with increased protectionism vis-à-vis foreign investments and with the rise in domestic fuel prices associated with the reductions in fuel subsidy. The other decomposition shows that the food prices component of the RER is the one putting most upward pressure to the RER. Indonesia had the second highest average annual relative food price inflation over 2003-2015 and this food price inflation contributed to the slower pace of the overall RER depreciation over 2014-15. This confirms the role of protectionist food trade policy in undermining the price competitiveness of Indonesian exports. On the other hand our analysis suggests that changes in the relative unit labor cost have not played any significant role in the slow rate of the recent Indonesian RER depreciation. 1 1. Introduction One of the most prominent effects of commodity booms is the appreciation of the Real Exchange Rate (RER), the so-called ‘Dutch Disease’ effect. Resources flow to the booming commodity sector of the economy and that typically stimulates spending on tradable goods, which increase imports, and on non tradable sectors, whose prices increase. This price increase generates an appreciation of the RER, which can also be expressed as the ratio of the price of non tradables over tradables. As we document in this paper Indonesia is no exception. Irrespective of the RER definition used, Indonesian RER exhibited a strong appreciation up until mid-2012, in line with other commodity dependent middle-income countries such as Brazil and South Africa. This RER appreciation is bad news for the non-booming tradable sector and in turn for economic growth (Rodrik, 2008). Firms source all of the non tradable inputs and much of their goods’ inputs domestically and if their price rises faster than the international price at which they can sell their goods, their relative competitive position deteriorates. On the other hand by increasing firms’ sales and profits an undervalued RER increases the rate of return to capital employed in the production of tradable goods. If this increase is sufficiently large this can overcome the institutional problems that typically affect the tradable sector in developing countries, thus increasing the rate of investments in that sector (Rodrik, 2008 and Rapetti, 2013).1 Indeed RER appreciation is one of the main negative effects of aid – which is an inflow of resources similar to that of a commodity boom - on manufacturing export and growth (Rajan and Subramanian, 2011). Consistently with this view Indonesian manufacturing exports have declined in relative terms during the commodity boom period and the RER appreciation has been a key factor behind this decline (World Bank, 2014). Indeed prices of non-tradable goods have increased threefold in the period 2000-10, partly as the commodity boom increased demand for domestic non tradable inputs, where competition is limited (Ghosh, Rahardja and Varela, 2012). The RER appreciation tends to be particularly detrimental for sectors likely to be key to ensure Indonesian sustained growth going forward, i.e. manufacturing and modern services (World Bank, 2014). Rajan and Subramanian (2011) indirectly show that labor intensive manufacturing as wearing apparel, footwear, wood and leather products specialization are among the most harmed manufacturing sectors from an appreciation of the RER. Similarly sectors with increasing returns to scale also typically suffer from a RER appreciation, as the associated reduction in sales harm their productivity (Pyun and Choi, 2015). Examples 1 RER depreciation will also increase the cost of imported intermediate goods and may cause increasing inflation and lowering growth. If the substitutes of these inputs are not available in the country, then depreciation may deteriorate the trade balance (Gylfason and Schmid, 1983, Wijnbergen, 1989). 2 of such sectors include machinery equipment, electrical equipment and transport equipment, which empirical evidence has shown to be highly subject to RER movements (Fouquin et al., 2001). And the RER appreciation reduces the exports of modern services even more than merchandise exports (Eichgreen and Gupta, 2012). These detrimental effects of the RER appreciation have contributed to the emergence of a ‘wrong’ type of specialization in Indonesia, one skewed towards commodity based and low-productive non tradable services sectors.2 This has contributed to low labor productivity and sluggish Total Factor Productivity (TFP) growth, as manufacturing and the modern service sectors have larger scope for technological advance. By being more exposed to international markets, these sectors allow an economy to absorb best practices from abroad (Rajan and Subramanian, 2011). Indeed virtually all countries that have had a sustained period of growth in the post-war period have seen a large increase in their share of manufacturing and manufacturing exports (Jones and Olken, 2005). Despite the importance of the RER movements for Indonesian sectoral specialization and future growth prospects, not much analysis has focused on the issue. Moreover the analyses mainly use the real effective exchange rate (REER) to measure the RER. This is an imperfect proxy of the RER as it compares Indonesian RER to that of its main trading partners. However they are not necessarily Indonesian main competitors. For example countries like Vietnam, Malaysia and the Philippines do not feature prominently in the REER but are important competitors in third markets for Indonesian firms. In addition to the best of our knowledge little recent analysis exists on the drivers of RER movements in Indonesia. This paper aims to help fill these gaps by conducting an investigation of the evolution of the bilateral price based RER (CPI RER) between Indonesia and its main comparators in the period 2002-2015. These countries include the US (as the benchmark), China, the Philippines, Vietnam, Malaysia, Thailand, India, Brazil and South Africa. In addition we examine the drivers of the RER movements by performing two novel decompositions of the bilateral RER to. The results show that Indonesia’s CPI RER exhibited strong real appreciation trend during the commodity price hike through mid-2012. The magnitude of real appreciation was stronger relative to other Asian comparator countries and lower only to commodity exporters Brazil and South Africa. While all Asian countries experienced nominal exchange rate appreciation between January 2002 and June 2008, of which Indonesia’s Rupiah appreciated the least, Indonesia’s relative price inflation was largest among the Asian countries and contributed to higher RER appreciation over this period. 2 Manufacturing went from 28 percent of GDP in 2000 to 25.5 percent in 2014. 3 Indonesia’s CPI RER depreciated after 2011 vis-a-vis all comparators execpt the commodity exporters mainly driven by nominal depreciation. However the pace of the recent (2014-15) CPI RER depreciation is slower compared to earlier 2011-13 period and is much attenuated relative to the size of the nominal depreciation. The food-non food prices decomposition analyses show that the relative food prices in general contributed to Indonesian RER appreciation vis-à-vis the comparators throughout the entire period. Indonesia had the second highest average annual relative food price inflation over 2003-2015. The relative food price inflation in Indonesia remained high over the last three years (2013-15, 5.7%, trailing only Brazil and India), contributing to slower pace of the overall RER depreciation in recent years. This confirms the role of protectionist food trade policy in undermining the price competitiveness of Indonesian exports. The other RER decomposition shows that local retail prices rose faster than border prices (relative to the US and Asian peers) since the second quarter of 2014, which counterbalanced stronger depreciation of the tradable RER and implied a slower rate of depreciation of the CPI RER over this period. We hypothesize that this is due to increases in tariffs and non tariff barriers as well as in distribution costs, which in turn is consistent with increased protectionism vis-à-vis foreign investments and with the rise in domestic fuel prices associated with the reductions in fuel subsidy. On the other hand relative prices of non-tradable goods show small depreciation vis-à-vis Asian currencies and USD over this period (stronger relative to the USD in 2015). The changes in the relative unit labor cost are associated with the changes in tradable RER, but explain only a small part of the movements in the distribution wedge, suggesting their limited role in slower rate of the recent CPI RER depreciation. This finding suggests that wage adjustments may not be needed for Indonesia to regain price competitiveness. The lack of further depreciation in these post-commodity boom period is a key reason behind the inability of firms to regain the competitiveness lost during the commodity boom period. That creates huge challenges for firms to try to expand labor demand. Our results suggest that raising barriers to goods’ trade and to investments in serv ices sectors key to goods’ production, such as warehousing and distribution, have reduced the extent of the depreciation thus contributing to the sluggish performance of the non commodity export sector. Addressing these barriers would be crucial for a Dutch Disease reversal to happen in Indonesia allowing these sectors to drive the growth of the economy. 2. Real Exchange Rate Methodology The real exchange rate is the most commonly used indicator of international price and cost competitiveness and one of the key variables in determining country’s current account. The 4 real exchange rate is typically computed as a geometric weighted average of bilateral real exchange rates between the home country and a set (basket) of other countries. The real effective exchange rate index E for country i constructed in such way is defined as: E = ∏ ≠ ( ) (1) where j refers to other country, P’s are respective price deflators and Ri and R j are bilateral nominal exchange rates of country i and j against the U.S. dollar (measured in U.S. dollar per local currency). When constructing the REER index several methodological issues need to be addressed. This includes: (i) the selection of countries to be included in the basket and the size of the basket (n); (ii) the method for computing the weights for each country (Wij); (iii) the choice of price deflator to obtain the real counterparts of the nominal effective exchange rate. The common choices (BIS, Bayoumi et al, 2005) in each domain are: (i) the largest trading partners where the size of the basket varies per country; (ii) the weights are computed using bilateral trade intensity with/no taking into account third- country competition; (iii) consumer price indices (CPI) or, less frequently depending on data availability, unit labor cost (ULC). While providing a useful summary of country’s overall price competitiveness, the REER indices do not allow for direct comparisons across the countries as basket sizes and their compositions may significantly differ, both cross-sectionally and over time. In this report we therefore focus on Indonesia’s bilateral real exchange rates and study their evolution vis-à-vis peer countries. The results of the analysis should thus be viewed as complementary to the standard usage of the REER and focused more on mapping the differences in dynamics of Indonesia’s real exchange rate and its components relative to the peer countries. Bilateral real exchange rate is defined in equation (2) as the ratio of the price level in home country (P) to the price level in the foreign country ( P*) in terms of the same currency where R represents the number of units of foreign currency for one unit of home currency. Thus, an increase in the RER implies an appreciation of the home currency with respect to the foreign currency in real terms. E = (2) ∗ Analogously to the REER, bilateral real exchange rate is typically computed using aggregate CPI indices. Aggregate CPI indices however include the prices of tradable and non-tradable goods and services which may show different dynamics, both in theory and practice. Traditional real exchange rate theory dating back to Cassel (1918), Balassa (1964) and Samuelson (1964) assumes that the relative prices of tradable goods between the countries are stable as all deviations from the law of one price are eliminated in the short- 5 term by arbitrage activities. The bulk of the changes in the RER are therefore associated with the movements in the relative prices of non-tradable goods across countries since their prices are domestically determined and not subject to arbitrage opportunities. Following the intuition of the traditional theory, the real exchange rate can alternatively be defined as the relative price of non-tradable to tradable goods in two countries. The new open economy macro models (starting from Betts and Devereux, 2000, Chari et al, 2002) on the other hand allow for persistent deviations from the law of one price driven by different types of nominal rigidities. In empirical study of the bilateral exchange rates between the United States and a number of OECD countries Engel (1999) documents that majority of the overall fluctuations in the CPI real exchange rates are in fact associated with the changes in the real exchange rates of tradable goods, while the fluctuations in the relative prices of non-tradable to tradable goods fail to explain much of the variation of the overall RER. Following this stream of the literature, the real exchange rate can be defined solely in terms of the relative price of tradable goods. The key element in contrasting results of two strands of the empirical literature is the definition of the tradable goods prices. Engel (1999) applies tradable goods prices at the retail level. Burstein et al (2005, 2006) and Betts and Kehoe (2006, 2008) subsequently argue that the usage of retail prices may introduce measurement bias due to significant non- traded components in them related to, inter alia, transportation and marketing costs. The tradable goods prices defined in this way thus can overestimate the importance of tradable goods for the RER dynamics. Burstein et al (2005, 2006) instead define tradable prices using aggregate import and export prices at-the-dock and allocate the remaining part of the retail price to non-tradable component. Similarly, Betts and Kehoe (2006, 2008) define tradable prices using producer price indices (PPI) and sectoral gross output deflators. The latter papers indeed find stronger importance of the relative price of non-tradable goods compared to Engel (1999), though still not dominant in explaining the movements in the overall real exchange rate. Bache et al (2013) recently offers a promising connection between the two competing sets of empirical results. They introduce the concept of relative distribution wedge as the wedge between the retail prices of traded goods and at the dock prices of these goods in two countries. The relative distribution wedge measures differences in local distribution costs and time-varying local mark-ups (in part related to local price stickiness), which also differs from traditional measures of non-tradable goods prices. They show that the measure explains close to 50 percent of overall variation in bilateral real exchange rates of 9 OECD countries and provides alternative explanation of earlier conflicting findings in the literature. In line with the existing literature, in this report we construct the overall CPI based bilateral RER vis-à-vis USD and the peer countries’ currencies and decompose it to three different components (measures) of RER. In particular, following Engel (1999), Betts and Kehoe (2006) and Bache et al (2013) the overall CPI RER is defined as: 6 ∗ /∗ / E= ∗ ∗ ∗ / / or equivalently: / / E= ∗ ∗ ∗ ∗ ∗ / / Where P, P* are again the overall CPI in home and foreign country respectively; PRT are the retail process of the tradable goods and PT are at-the-dock (border) prices of these goods. Taking logs, the CPI-based RER can be decomposed in three terms: e = + + ∗ ) = ( − ) − (∗ − (3) ∗ = + − (4) ∗ ∗) = ( − ) − ( − (5) The first term (equation 3) is the ratio of relative prices of non-tradable to tradable goods which provides a traditional measure of the RER. The second term (equation 4) is the real exchange rate of tradable goods measured using at-the-dock prices. The last term (equation 5) is the measure of the distribution wedge. In the remainder of the paper we compute and compare all four measures (equations 2-5) of the RER. In addition, following the same steps as above we decompose the overall log CPI RER in the sum of the log nominal exchange rate, the log of relative food prices and the log of relative non-food prices. 3. Data We use monthly and quarterly data to compute aggregate CPI based RER vis-à-vis United States (equation 2) for Indonesia and its main competitors in the global manufacturing market. We include the following eight countries: Brazil, China, India, Malaysia, the Philippines, South Africa, Thailand and Vietnam. Monthly data on the nominal exchange rates and local CPI indices (monthly averages) over the January 2002-December 2015 period are obtained from the IMF’s International Financial Statistics (IFS) database. Due to data availability we construct the components (alternative measures) of the RER (equations 3-5) at quarterly frequency, starting where available in the first quarter of 2002. The nominal exchange rates (quarterly averages) again come from the IFS. Local CPI indices and the retail prices of tradable goods are calculated using data from local statistical offices. Since the retail prices of tradable goods are not directly available, we use CPI data 7 on “goods” and “services”, where “goods” are regarded as traded goods. In particular, from each CPI we extract series classified as goods by UN COICOP methodology3. If the weights for tradables are not directly available from the source data, we calculate them in two steps. Take for example Indonesian CPI group “Housing”. It consists of 4 subcategories: 1) Fuel, Electricity and Water; 2) Household Equipment; 3) Cost of Housing; and 4) Household Operations. We have the time series of prices for the broad category and four subcategories but we don’t know the weights each of the four categories has in the broad index. If we want to replicate actual broad price index we need to back out the weights such that our estimates match the actual movements in the broad index. Once we have the estimates of the weights for each subcategory we can select only the subcategories which are related to tradables. To back out the weights in the first step we use the fact that the weights do not change frequently. Since we have four subcategories, hence four unknown weights we need to set up a system of four linear equations. The first equation is just that four weights sum to one. The second equation is that “Housing” index in January 2002 is the weighted average of four subcategories. The third equation repeats second equation for February 2002. The fourth does that for March 2002. The weights for four categories are obtained as the solutions to this system of four equations. Then, in the second step we assign the computed weights to tradables and nontradables. In the present example, the first two categories are selected as goods and their weights are saved. We repeat the steps over time for the same category and all steps for all other categories for which we don’t have information on the actual weight s. The final CPI index for tradables is calculated by normalizing all weights for the series defined as goods so that all the weights sum to 1. Table 1 shows the share of goods in the total CPI for the countries under analysis. Given that the share of services in consumption tends to increase with income, the share of goods in CPI is higher in all countries relative to the United States and at comparable levels between the countries. Finally, at-the-dock prices of tradable goods are calculated as the weighted average of aggregate import price indices and export price indices where the weights are computed using the country level trade value data obtained from IFS.4 The data on import and export price indices is collected from local statistical offices and at the dock prices are computed as: = (1 − )( ) + ( ) 3 That is the overall CPI without services, which are defined as: actual rentals paid by tenants; imputed rentals for owner occupiers; services for the repair and maintenance of a dwelling (plumbers and electricians); water supply; other services relating to the dwelling n.e.c.; domestic services and household services; medical services; dental services; hospital services; maintenance and repair of personal transport equipment; other services in respect of personal transport equipment; transport services; postal services; telephone and telefax services; recreational and sporting services; cultural services; games of chance; education; restaurants and hotels; accommodation services; social protection services; package holiday; insurance; financial services n.e.c.; other services n.e.c. 4 These are aggregate indices computed by the local statistical offices. 8 where the weight a represents the average share of export value in total value of foreign trade, IPI is the aggregate import price index (based on CIF prices) and EXI is the aggregate export price index (based on FOB prices). Table 2 reports the share of export values for countries in our sample, where we see some but moderate dispersion across the countries. Data for food and non-food price indices is obtained from local statistical offices. The food price index is the weighted average of the raw food, the processed food and the beverages indices. 4. Results We start the analysis with the traditional measure of the real exchange rate – the CPI REER obtained from the BIS database ( 9 Figure 1). Indonesia’s CPI REER broadly follows the commodity price trends. The REER appreciated prior to mid 2003 and following short reversal in 2004 and mid 2005, it continued appreciating more stongly until August 2008 in line with the first spike in commodity prices. The beginning of the global financial crises, the fall in commodity prices and strong nominal depreciation of the Rupiah vis-à-vis USD and other major currencies over the subsequent months depreciated the REER. Another period of REER appreciation started quickly after, from February 2009, and continued until mid 2011 throughout the second spike in commodity prices. The REER then started to depreciate until the end of 2013. Over the 2014-15 period, the REER appears to be rather sticky and gradually appreciating contrary to other commodity exporters. We next move to the analysis of the bilateral RER vis-à-vis USD for Indonesia and comparator countries over the same period (Figure 2). Indonesia’s CPI RER displays similar behavior to the REER until recently with several interchanging periods of real depreciations and appreciations. The last two years however differ. While the REER suggests small real exchange rate appreciation, strong nominal depreciation of Rupiah vis- à-vis USD over 2013-15 has resulted in the RER depreciation, albeit the strength of depreciation is weaker relative to the 2011-13 period. That is because the REER compares Indonesia to various other trading partners, whose real prices have declined relatively to Indonesia during this period. Indeed compared to the peer countries, Indonesia’s RER displays overall appreciation trend over the last 13 years of similar magnitude relative to China, India, the Philippines and Vietnam while the monthly movements appear to be more volatile relative to them. On the other hand, Indonesia’s RER exhibits significantly lower volatility compared to Brasil and South Africa, while higher overall depreciation. Thailand’s and Malaysia’s RERs show lower and less volatile RER appreciation over the same period with actual real depreciation of Malysia’s Ringgit in 2015 relative to its 2002 level. We next divide the entire period in two sub-periods: the period of the commodity price boom until mid 2008 and the period thereafter which saw a significant fall in commodity prices. The second subperiod also includes a boom reversal over 2009-2012, followed by another strong fall in commodity prices. In the first subperiod Indonesia’s RER exhibits strong real appreciation trend (Figure 3). The magnitude of real appreciation is stronger relative to other Asian comparator countries and lower only to other commodity driven economies, Brasil and South Africa. Figure 4 decomposes changes in the RER into nominal exchange rate and relative price contributions. While all Asian countries experienced nominal exchange rate appreciation between January 2002 and June 2008, of which Indonesia’s Rupiah appreciated the least, Indonesia’s relative price inflation was largest among the Asian countries and contributed to higher RER appreciation over this period. The behavior of Indonesia’s CPI RER is markedly different over the second subperiod. The dynamics are again more in line with 10 Brasilian Real and South African Rand RERs changes, relative to other Asian countries. In particular, the Rupiah-Dollar RER depreciated during the commodity bust periods (2008, 2011-onwards), while it appreciated during the commodity price rebound in between. The magnitude of the recent RER depreciation is lower relative to the two commodity-based countries, more notably over 2014-15. Other Asian countries experienced less volatile RERs. Driven by nominal exchange rate depreciation with low inflation differential (Figure 7), Malaysia’s and Thailand’s RER depreciated more strongly over the most recent period. In particular, despite Rupiah’s nominal depreciation against the Dollar of 33% between June 2008-December 2015 which was stronger compared to the Ringgit, higher domestic CPI inflation relative to Malaysia contributed to lower Dollar RER depreciation in Indonesia. On the other end, China, India, the Philippines and Vietnam contiuned experiencing RER apreciation trend vis-à-vis dollar and consequently the Rupiah over the second subperiod. Overall, the CPI RER analysis suggests that Indonesia’s RER behavior is strongly influenced by the global commodity price movements. It became decoupled from several other Asian countries over the post 2008 period (India, China, the Philippines and Vietnam), as the most recent nominal depreciation of Rupiah vis-à-vis US dollar have contributed to RER depreciation against the dollar and currencies of these countries. This is different to movements in the trade weighted REERs which all (Indonesia, India, China, the Philippines and Vietnam) exhibit a small appreciation over the most recent period. The pace of the recent CPI-measured depreciation is however slower compared to earlier 2011- 13 period and relative to the size of the nominal depreciation. The fact that Indonesia’s Dollar RER behavior broadly is closer to commodity driven economies (Brazil and South Africa) than its Asian peers is in line with the relative importance of commodities in the foreign trade of these economies (Table 3). The average share of commodities (over 2010- 2014) in total trade of Indonesia (23%) is more similar to the high shares for Brazil and South Africa than the lower shares of the Asian peers (China 12%, Malaysia 15%, Thailand 16%). The commodity price movements are more strongly reflected in exports where their share is higher for Brazil, South Africa and Indonesia (53%, 38% and 30%). We next move to the analysis of the three components of the RER (non tradables, tradables and distribution wedge). Analyzing their dynamics provides deeper understanding of the observed changes in the CPI RERs and their implications for the manufacturing exports price competitiveness. Figure 5 reports estimates of the quarterly Dollar RER CPI and of its three components for countries and periods for which the data is available. Due to lack of data availability, India and Vietnam are not included in the analysis and the Chinese and Malaysian series start later. The reported RER measures are expressed in natural logartihms and re-scaled relative to the first period for which the data is available. In particular it shows annual changes in the Indonesia’s CPI RER vis-à-vis each comparator country, 11 decomposed into contributions from 3 components (ie, the numbers for 2003 report change in the RER in 2003 relative to 2002).5 The estimates for Indonesia show that until 2008 the bulk of observed CPI RER appreciation was driven by the tradable goods RER appreciation against most comparators in line with the large share of commodities in Indonesian trade. The exception was Brazil due to its even larger share of commodities in its trading basket. Indonesia’s tradable RER appreciated vis-à-vis other Asian countries until 2012 (2013 in case of China), the only exception is Thai Baht in 2009. However, since the beginning of the second commodity fall period (Q1 2012), Indonesia’s tradable RER has depreciated vis-a-vis majority of other Asian currencies (China, the Phillipines, Thailand), while appreciated with respect to Brazil (more) and South Africa (less). It has marginally appreciated vis-à-vis Malaysia over 2014-15. The Indonesian tradable goods RER vis-à-vis the US depreciated in early 2009, and after stronger appreciation between the second quarter of 2009 and the first quarter of 2012, it started depreciating thereafter. The pace of the depreciation slowed down in the first three quarters of 2013, but continued more strongly since then. In line with Engel (1999), the ratio of relative prices of non-tradable to tradable goods does not significantly contribute to the CPI RER movements. Indonesia’s non-tradable to tradable goods Dollar RER shows small, but persistent depreciation from 2008 onwards, similar in size to China and Thailand and larger compared to other comparator countries. The annual depreciation of non tradable to tradable ratio vis- à-vis the peer countries appears stronger with respect to the US, South Africa and Malaysia over the post 2012 period. The distribution wedge component depreciated vis-à-vis Brazil, China, Malaysia, the Philippines and Thailand over 2011-2014 (Brazil 2012-14). However, since the second half of 2014 (Figure A1) Indonesia’s local retail prices increased faster than the border prices compared to other Asian countries (the wedge appreciation). This resulted in the overall Rupiah’s RER appreciation (versus Malaysia, Thailand) and slower RER depreciation (versus China and the Philippines) in 2015. In line with Bache et al (2013) the distribution wedge RER explains some part of the overall CPI RER dynamics, with several periods of real depreciation and appreciation. In particular, the distribution wedge appreciates more strongly since the second quarter of 2014, which counterbalances stronger depreciation of the tradable RER and implies a slower depreciation of the CPI RER over this period. The movements in distribution wedge capture differences in local distribution costs, time-varying local mark-ups and tariff and non tariff measures. Thus the increase in this distribution wedge over 2014-15 may partly reflect the restrictions on foreign investments in warehouse and distribution services 5 Given that this RER decomposition is performed using different price data series than for the previous figures, we check the consistency between the two sets of bilateral RER series (figure 2 vs. figure 5). The comparison confirms that the two sets track each others very accurately (results available upon request). 12 imposed in 2014, the increasing use of trade restrictive measures in 2014 and the first half of 2015 (Figure 6)6, including the hike in tariffs on many consumer goods, and the rise in domestic fuel prices due to the elimination of fuel subsidies in 2014. Figure A2 in the appendix shows the same decomposition as in Figure 5 but with the border prices computed using only the import price component (excluding the export prices). This decomposition allows to exclude the direct effect of exported commodity prices on the tradable component of the RER. The changes (annual depreciation/appreciation of each component) match the findings from Figure 5 to a large degree; however the size of the component’s annual changes differs in some cases. The only qualitative difference appears in the case of the Rupiah-Renminbi RER where we find that distributional component didn’t appreciate in 2015, while at the same time the size of the tradable goods depreciation vis-à-vis the Renminbi is smaller if we use only the import prices. Smaller extent of tradable goods depreciation and retail price appreciation in 2015 (relative to Figure 5) is also observed with respect to other Asian countries. The observed differences between Figure A2 and Figure 5 may partly capture the differences in the relative importance of commodities in imports and overall trade. For example, in the case of the Rupiah-Renminbi RER, the commodity movements are more important for the Rupiah when we add the export prices, while more important for the Renminbi with import prices only. Next we investigate to what extent these changes are driven by the food versus the non food components of prices. To that end we decompose the bilateral CPI RER into the nominal bilateral exchange rate, relative food price and relative non-food price components. Figure 7 shows the contributions of each component to annual change in the CPI RER. The relative food prices persistently contributed to Dollar RER appreciation in Indonesia over the entire period. This is similar to all other comparator countries, although the size of the contribution varies across the countries and time. The average annual relative food price inflation over 2003-2015 in Indonesia was 5.4%, trailing Vietnam and slightly above Brazil and India. Other comparator countries had lower average relative food price changes, with minimum in Malaysia (1.2%). The relative food price inflation in Indonesia remained high over the last three years (2013-15, 5.7%, trailing only Brazil and India), contributing to slower pace of the overall RER depreciation. Relative non-food prices in Indonesia also rose over 2014-15 (for the first time after 2005), however the size appears to be small (0.1% and 0.5% annually). The importance of food prices in keeping upward pressure on the Indonesian RER is consistent with Marks (2015), who finds that between 2008 and 2015 the nominal rate of protection (NPR) has grown substantially across sectors, and in food crops in particular. 6 According to Global Trade Alert data during that period (2014-q2 and 2015-q4) Indonesia has been consistently in the world’s top-10 users of restrictive trade barriers except in 2014-q4 and 2015-q2. 13 Besides rice, which is a key driver of the increased food inflation, the NPR increased for several food products, such as sugar, meat, fruits, vegetables and wheat flour. Finally, we examine the association of the constructed RER series with the changes in the relative real unit labor costs. Given the data availability, quarterly series of unit labor cost are constructed using data for manufacturing sector only. The data ends in Q4 2014. Figure 8 shows dynamics of the real unit labor costs which exhibit several depreciation and appreciation episodes, with stronger real depreciation from the beginning of 2012. Table 3 shows that the changes in relative unit labor cost are associated with the changes in tradable RER as the correlation between the two series is equal to 0.66. The correlation with other RER measures is smaller. The correlation between the relative ULC and distribution wedge RER is positive and moderate (0.26), which suggests that variations in the labor distribution costs (proxied by the relative ULC as in Bache et al, 2013), can explain only a small part of the movements in the distribution wedge. Overall, the analysis of the three RER components implies that the bulk of observed CPI RER movements are driven by the tradable goods RER dynamics, making Indonesia’s RER more closer to Brazil and South Africa relative to Asian countries. Indonesia’s tradable RER and relative price of non-tradable to tradable goods depreciated over the most recent (post 2012) period vis-à-vis Dollar and majority of Asian currencies. Appreciation of the distribution wedge component explains slower CPI RER depreciation over this period. 5. Conclusions We have documented that Indonesia confirms the pattern of commodity dependent countries that commodity price hikes induce the appreciation of the RER. That is true vis- à-vis all comparators except other commodity dependent countries. In line with other empirical evidence, the bulk of this appreciation was driven by the increase in tradable prices due to the importance of commodities in the Indonesia trading basket. This increase in tradable prices relative to those in comparator countries has been roughly matched by the increase in non tradable prices, so that their ratio changed in a similar way to comparator countries. Thus it has not contributed much to the changes in the RER throughout the period. This RER appreciation played a major role in the relative decline of Indonesian non commodity exports, manufacturing in particular, during that period. Importantly, the RER appreciation has only been partly reversed during the dramatic decline in commodity prices over the past 4-5 years, and it has been more sticky downwards than the other commodity exporting countries in the analysis. It has also been more sticky relative to the extent of the nominal depreciation of the currency in 2013-15. This stickiness 14 in 2014-15 is mainly due to local retail prices rising faster than border prices (relative to the US and Asian peers). We hypothesize that this is due to increases in tariffs and non tariff barriers as well as in distribution costs, which in turn is consistent with increased protectionism vis-à-vis foreign investments and with the rise in domestic fuel prices associated with the reductions in fuel subsidy. We also show that the food prices component of the RER is the one putting most upward pressure to the RER. Indonesia had the second highest average annual relative food price inflation over 2003-2015 and this food price inflation contributed to the slower pace of the overall RER depreciation over 2014-15. This confirms the important role of protectionist food trade policy in undermining the price competitiveness of Indonesian exports. On the other hand our analysis suggests that changes in the relative unit labor cost have not played any significant role in the slow rate of the recent Indonesian RER depreciation. The lack of further depreciation in these post-commodity boom period is a key reason behind the inability of firms in tradable non commodity sectors to regain the competitiveness lost during the commodity boom period. That deprives these firms from a natural competitive boost when trying to expand labor demand. Our results suggest at least two policy implications to help firms’ price competitiveness in Indonesia. First both tariff and non tariff barriers to goods’ trade, esepcially food, should be reduced. Indeed recent estimates (Marks, 2015) suggest very high and increasing nominal and effective rates of protection of Indonesian goods, especially in the food sector. For example rice price is estimated to be 68.7% more expensive than it would otherwise be under free trade regime. All the trade measures increase the cost of living in Indonesia by 7.4%. Second, barriers to investments in services sectors key to goods’ production, such as warehousing and distribution, should be reduced. The recent introduction of measures to limit foreign equity participation should be reversed, a suggestion much in line with other empirical evidence (Duggan et al., 2015). Similarly measures that have increased minimum capital requirements for logistics service providers are also not conducive to regaining price competitiveness for firms. Indeed Indonesia is the most restrictive among the 42 countries surveyed by the OECD in several logistics services, including freight forwarding, maritime transport and distribution. 7 Addressing these barriers would be crucial for a Dutch Disease reversal to happen in Indonesia allowing these sectors to drive the growth of the economy. 7 OECD Services Trade Restrictiveness Index 2015. 15 Figure 1: Real effective exchange rates, January 2002-December 2015: 130 110 90 70 50 30 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Indonesia Brazil China India 130 110 90 70 50 30 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Malaysia Philippines South Africa Thailand Source: BIS . 16 Figure 2: CPI Real exchange rates vis-à-vis USD, January 2002-December 2015: Indonesia Brazil China 240 240 240 220 220 220 200 200 200 180 180 180 160 142 160 160 136 140 140 113 140 120 120 120 100 100 100 80 80 80 60 60 60 jan.02 jan.09 dec.15 jan.02 jan.09 dec.15 jan.02 jan.09 dec.15 India Malaysia 240 240 240 220 220 220 200 200 200 180 180 180 160 145 160 160 141 140 140 140 120 120 93 120 100 100 100 80 80 80 60 60 60 jan.02 jan.09 dec.15 jan.02 jan.09 dec.15 Philippines 240 240 240 220 220 220 200 200 200 180 180 180 153 160 160 160 126 128 140 140 140 120 120 120 100 100 100 80 80 80 60 60 60 Vietnam South Africa Thailand Source: Authors’ elaboration on the basis of data from IMF IFS 17 Figure 3: CPI Real exchange rates vis-à-vis USD, January 2002-June 2008: Indonesia Brazil China 220 220 220 200 200 185 200 180 180 180 155 160 160 160 140 140 140 116 120 120 120 100 100 100 80 80 80 60 60 60 jun.08 jan.02 apr.05 jun.08 jan.02 apr.05 jan.02 apr.05 jun.08 India Malaysia 220 220 220 200 200 200 180 180 180 160 160 160 140 127 140 128 140 114 120 120 120 100 100 100 80 80 80 60 60 60 jan.02 apr.05 jun.08 Philippines Jan-02 Apr-05 Jun-08 220 220 220 200 200 200 174 180 180 180 160 160 138 160 134 140 140 140 120 120 120 100 100 100 80 80 80 60 60 60 Vietnam South Africa Thailand 18 Figure 4: CPI Real exchange rates vis-à-vis USD, June 2008-December 2015: Indonesia Brazil China 130 130 130 117 120 120 120 110 110 110 100 92 100 100 90 90 90 80 80 80 70 61 70 70 60 60 60 50 50 50 dec.15 jun.08 mar.12 dec.15 jun.08 mar.12 jun.08 mar.12 dec.15 India Malaysia 130 130 130 114 120 110 120 120 110 110 110 100 100 100 90 90 81 90 80 80 80 70 70 70 60 60 60 50 50 50 Philippines jun.08 mar.12 dec.15 jun.08 mar.12 dec.15 130 130 130 114 120 120 120 110 110 110 100 100 92 100 90 90 90 80 72 80 80 70 70 70 60 60 60 50 50 50 Vietnam South Africa Thailand Source: Authors’ elaboration on the basis of data from IMF and IFS 19 Figure 5: RER decomposition, 2002-2015: Annual contributions to RER CPI change: tradable prices based on export and import indices RER USA_INDONESIA RERs BRASIL_INDONESIA RER CHINA_INDONESIA 0.20 0.40 0.20 0.15 0.30 0.15 0.10 0.10 0.20 0.05 0.05 0.10 0.00 0.00 0.00 -0.05 -0.05 -0.10 -0.10 -0.10 -0.20 -0.15 -0.15 -0.30 -0.20 2003 2005 2007 2009 2011 2013 2015 2003 2005 2007 2009 2011 2013 2015 2003 2005 2007 2009 2011 2013 2015 rer_n rer_t rer_d rer _cpi rer_n rer_t rer_d rer _cpi rer_n rer_t rer_d rer _cpi RER MALAYSIA_INDONESIA RER PHILIPPINES_INDONESIA RER SOUTH AFRICA_INDONESIA 0.15 0.20 0.25 0.15 0.20 0.10 0.15 0.10 0.10 0.05 0.05 0.05 0.00 0.00 0.00 -0.05 -0.05 -0.05 -0.10 -0.10 -0.15 -0.15 -0.10 -0.20 -0.20 -0.25 -0.15 -0.25 -0.30 2003 2005 2007 2009 2011 2013 2015 2003 2005 2007 2009 2011 2013 2015 2003 2005 2007 2009 2011 2013 2015 rer_n rer_t rer_d rer _cpi rer_n rer_t rer_d rer _cpi rer_n rer_t rer_d rer _cpi 20 RER THAILAND_INDONESIA 0.20 0.15 0.10 0.05 0.00 -0.05 -0.10 -0.15 2003 2005 2007 2009 2011 2013 2015 rer_n rer_t rer_d rer _cpi Source: IMF IFS, local statistical offices and author’s calculations Figure 6: Indonesia’s liberalizing and restricting trade measures by quarter Source: Bank staff calculations based on Global Trade Alert (accessed 25 April 2016) 21 Figure 7: RER vis-à-vis comparatos decomposition, food vs. non food prices 2002-2015: Annual contributions: Indonesia-US contribution to y-o-y change of contribution to y-o-y change of 0.40 0.30 rer_bra_indo 0.30 rer_chi_indo 0.20 0.20 0.10 0.10 0.00 0.00 -0.10 -0.10 -0.20 -0.20 -0.40 -0.30 -0.30 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 ner cpi_food cpi_nonfood rer ner cpi_food ner cpi_food cpi_nonfood reer_cpi contribution to y-o-y change of contribution to y-o-y change of contribution to y-o-y change of 0.30 rer_ind_indo 0.40 rer_mal_indo 0.30 rer_phi_indo 0.20 0.20 0.20 0.10 0.10 0.00 0.00 0.00 -0.10 -0.10 -0.20 -0.20 -0.20 -0.30 -0.40 -0.30 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 ner cpi_food cpi_nonfood reer_cpi ner cpi_food cpi_nonfood reer_cpi ner cpi_food cpi_nonfood reer_cpi 22 contribution to y-o-y change of contribution to y-o-y change of contribution to y-o-y change of 0.30 rer_saf_indo 0.30 rer_tha_indo 0.30 rer_vie_indo 0.20 0.20 0.20 0.10 0.10 0.10 0.00 0.00 0.00 -0.10 -0.10 -0.10 -0.20 -0.20 -0.20 -0.30 -0.30 -0.30 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 ner cpi_food cpi_nonfood reer_cpi ner cpi_food cpi_nonfood reer_cpi ner cpi_food cpi_nonfood reer_cpi Source: IMF IFS, local statistical offices and author’s calculations Figure 8: Relative Unit Labor Cost, 2002Q1-2015Q4: 130 0.8 120 0.7 110 0.6 100 0.5 90 0.4 80 0.3 70 0.2 60 0.1 50 0 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 Note: left chart is index (2010=100), the right is in logs scaled to the first period Source: IMF IFS, local statistical offices and author’s calculations 23 Table 1: The share of goods in CPI (period average): Brazil 60.4 China 59.6 Indonesia 56.4 Malaysia 53.3 Philippines 60.4 South Africa 56.3 Thailand 57.2 USA 40.3 Source: Local statistical offices and author’s calculations. Table 2: The share of exports in overall trade (average over the period) Brazil 52.3 China 53.6 Indonesia 57.4 Malaysia 54.6 Philippines 46.8 South Africa 46.1 Thailand 50.4 USA 38.0 Source: IMF IFS Table 3: The share ofcommodities in overall trade (average over 2010-2014): Imports Exports Imports and Exports Indonesia 0.15 0.30 0.23 Brazil 0.09 0.53 0.32 China 0.21 0.05 0.12 Malaysia 0.16 0.16 0.16 South Africa 0.10 0.38 0.24 Thailand 0.10 0.20 0.15 Source: UN COMTRADE Table 4: Correlations between variables: Non-tradables RER Tradables RER Distribution wedge ULC -0.17 0.66 0.26 Source: Author’s calculations. 24 References Balassa, B., 1964: “The purchasing power parity doctrine: A reappraisal”. Journal of Political Economy 72(6), 584-596. Bayoumi, T., Lee, S., and J. Jayanthi (2005): "New Rates from New Weights". IMF Working Papers, IMF Working Paper 05/99. Bache, I.W., Sveen, T., and K.N. Torstensen (2013): “Revisiting the importance of non-tradable goods’ prices in cyclical real exchange rate fluctuations”. European Economic Review 57, 98–107 Betts, C., Devereux, M.B, 2000: “Exchange rate dynamics in a model of pricing-to-market”. Journal of International Economics 50 (1), 215–244. Betts, C.M., Kehoe, T.J., 2006: “U.S. real exchange rate fluctuations and relative price fluctuations ”. Journal of Monetary Economics 53 (7), 1297–1326. Betts, C.M., Kehoe, T.J., 2008: “Real Exchange Rate Movements and the Relative Price of Non-traded Goods”. NBER Working Paper 14437. Burstein, A., Eichenbaum, M., Rebelo, S., 2005: “Large devaluations and the real exchange rate”. Journal of Political Economy 113 (4), 742–784. Burstein, A., Eichenbaum, M., Rebelo, S., 2006: “The importance of nontradable goods’ prices in cyclical real exchange rate fluctuations”. Japan and the World Economy 18 (3), 247–253. Cassel, G., 1918: “Abnormal deviations in international exchanges”. Economic Journal 28, 413–15. Chari, V., Kehoe, P.J., McGrattan, E.R., 2002: “Can sticky price models generate volatile and persistent real exchange rates?”. Review of Economic Studies 69 (3), 533–563. Eichengreen, B. and P. Gupta (2012). “The Real Exchange Rate and Export Growth: Are Services Different?”, NIPFP Working Paper 112. Engel, C., 1999: “Accounting for U.S. real exchange rate changes”. Journal of Political Economy 107 (3), 507–538. Fouquin M., M. Nanno, N. Laurence, S. Khalid and M.J. Malek, 2001. "Sector Sensitivity to Exchange Rate Fluctuations", Working Papers 2001-11, CEPII research center. Ghosh, S., S. Rahardja and G. Varela (2012). “How the Macroeconomic Environment and Investment Climate Have Affected the Manufacturing Sector”, Policy Note 3, The World Bank. Jones, B., and B. Olken (2005). “The Anatomy of Start-Stop Growth”, NBER Working paper, No. 11528. Pyun, J. H., & Choi, B. Y. (2015). Does Real Exchange Rate Depreciation Increase Firm Productivity? Analysis Using Korean Firm-Level Data. Korea Institute for International Economic Policy, Sejong. Rajan, R.G. and A. Subramanian (2011). “Aid, Dutch Disease, and Manufacturing Growth”, Journal of Development Economics, 94(1): 106-118. 25 Rapetti, M. (2013). “The real exchange rate and economic growth: some observations on the possible channels”, Working Paper No. 2013-11, University of Massachusetts, Department of Economics. Rodrik, D. (2008). “The real exchange rate and economic growth”, Brookings papers on economic activity, 2008(2), 365-412. Samuelson, P., 1964:“Theoretical notes on trade problems”. Review of Economics and Statistics 46(2), 145-154. World Bank (2014). “Indonesia: Avoiding the trap”, Development Policy Review, Jakarta: The World Bank 26 Figure A1: RER decomposition, 2002 Q1-2015 Q4: Quarterly contributions to RER CPI change: tradable prices based on export and import indices RER USA_INDONESIA RERs BRASIL_INDONESIA RER CHINA_INDONESIA 0.15 0.3 0.15 0.25 0.1 0.1 0.2 0.05 0.05 0.15 0.1 0 0 0.05 -0.05 -0.05 0 -0.1 -0.1 -0.05 -0.15 -0.1 -0.15 -0.2 -0.15 -0.2 -0.2 -0.25 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 rer_n rer_t rer_d rer_n rer_t rer_d rer_n rer_t rer_d RER MALAYSIA_INDONESIA RER PHILIPPINES_INDONESIA RER SOUTH AFRICA_INDONESIA 0.15 0.2 0.15 0.1 0.15 0.1 0.1 0.05 0.05 0.05 0 0 0 -0.05 -0.05 -0.05 -0.1 -0.1 -0.15 -0.1 -0.15 -0.2 -0.2 -0.15 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 2002 2004 2006 2008 2010 2012 2014 rer_n rer_t rer_d rer_n rer_t rer_d rer_n rer_t rer_d 27 RER THAILAND_INDONESIA 0.15 0.1 0.05 0 -0.05 -0.1 -0.15 -0.2 -0.25 2002 2004 2006 2008 2010 2012 2014 rer_n rer_t rer_d Source: IMF IFS, local statistical offices and author’s calculations 28 Figure A2: RER decomposition, 2002-2015: Annual contributions to RER CPI change: tradable prices based on import indices RER USA_INDONESIA RER BRASIL_INDONESIA RER CHINA_INDONESIA 0.20 0.40 0.30 0.15 0.30 0.20 0.20 0.10 0.10 0.10 0.05 0.00 0.00 0.00 -0.10 -0.10 -0.05 -0.20 -0.20 -0.10 -0.30 -0.30 2003 2005 2007 2009 2011 2013 2015 2003 2005 2007 2009 2011 2013 2015 2003 2005 2007 2009 2011 2013 2015 rer_n rer_t rer_d rer _cpi rer_n rer_t rer_d rer _cpi rer_n rer_t rer_d rer _cpi RER MALAYSIA_INDONESIA RER PHILIPPINES_INDONESIA RER SOUTH AFRICA_INDONESIA 0.15 0.20 0.25 0.15 0.20 0.10 0.10 0.15 0.05 0.05 0.10 0.05 0.00 0.00 0.00 -0.05 -0.05 -0.05 -0.10 -0.10 -0.10 -0.15 -0.15 -0.20 -0.20 -0.15 -0.25 -0.25 -0.20 -0.30 -0.30 2003 2005 2007 2009 2011 2013 2015 2003 2005 2007 2009 2011 2013 2015 2003 2005 2007 2009 2011 2013 2015 rer_n rer_t rer_d rer _cpi rer_n rer_t rer_d rer _cpi rer_n rer_t rer_d rer _cpi 29 RER THAILAND_INDONESIA 0.15 0.10 0.05 0.00 -0.05 -0.10 -0.15 2003 2005 2007 2009 2011 2013 2015 rer_n rer_t rer_d rer _cpi Source: IMF IFS, local statistical offices and author’s calculations 30